INTRODUCTION TO ACCELERATORS

Transcription

1 INTRODUCTION TO ACCELERATORS (Presented in 2 lectures) CAS Granada, October 2012 P.J. Bryant CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 1

2 Contents Comment on accelerators Pre-accelerator era The main history line A second history line And another history line, but fainter Classification by Maxwell Status in 1940 After 1940 in a nutshell Classification of accelerators Where to next? A closer look at cyclotrons Recognising synchrotron lattices The FFAG. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 2

3 Comment on accelerators Modern accelerators can accelerate particles to speeds very close to that of light. At low energies, the velocity of the particle increases with the square root of the kinetic energy (Newton). At relativistic energies, the velocity increases very slowly asymptotically approaching that of light (Einstein). It is as if the velocity of the particle saturates. One can pour more and more energy into the particle, giving it a shorter De Broglie wavelength so that it probes deeper into the sub-atomic world. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 3

4 What s in the name? What does special relativity tell us, e.g. for an electron? Energy 1 MeV 1 GeV β = v/c γ = m/m Yes, the speed increases, but not as spectacularly as the mass. In fact, it would be more correct to speak of the momentum (mv) increasing. Ginzton, Hansen and Kennedy* suggested, Ponderator or Mass Agrandiser, but this did not become fashionable and we are left with Accelerator. * Rev. Sci. Instr., Vol.19, No.2, Feb CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 4

5 Pre-accelerator era <100 kev electrons from Wimshurst-type machines: 1895 Lenard electron scattering on gases (Nobel Prize 1905 for work on cathode rays) Franck and Hertz excited electron shells by electron bombardment. Few MeV from natural alpha particles: 1906 Rutherford bombards mica sheet with natural alphas Rutherford induces a nuclear reaction with natural alphas. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 5

6 Build-your-own Wimshurst machine (1903) 100 years ago physics experimentation was very popular with the general public who often built their own equipment. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 6

7 A commercial Wimshurst-type machine CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 7

8 The main history line Rutherford believes that he needs a source of many MeV to continue his research on the nucleus. This is far beyond the electrostatic machines then existing, but in 1928 Gamov predicts tunneling and perhaps 500 kev would suffice??? and so the first accelerator was built for physics research: 1928 Cockcroft & Walton start designing an 800 kev generator encouraged by Rutherford the generator reaches 700 kev and Cockcroft & Walton split the lithium atom with only 400 kev protons. They received the Nobel Prize in CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 8

9 The players Ernest Rutherford: Born 30/8/1871, in Nelson, New Zealand. Died Professor of physics at McGill University, Montréal ( ). Professor of physics at University of Manchester, UK ( ). Professor of experimental physics and Director of the Cavendish Laboratory, University of Cambridge. Sir John Douglas Cockcroft: Born 27/5/1897, Todmorden, UK. Died Ernest Thomas Sinton Walton: Born 6/10/1903, Ireland. Died 25/6/1995. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 9

10 Cockcroft & Walton s generator CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 10

11 Cockcroft-Walton generators became standard equipment The Cockcroft-Walton generator supplying the ion source which injected protons into NIMROD, the 7 GeV synchrotron at Rutherford laboratory. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 11

12 Van de Graaff, a competitor DC voltage generator: Van de Graaff was an American Rhodes scholar in Oxford, UK in 1928 when he became aware of the need for a high-voltage generator. His first machine reached 1.5 MV in Princeton, USA, in the early 1930s. These generators typically operate at 10 MV and provide stable low-momentum spread beams. [Robert Van de Graaff 20/12/ ] CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 12

13 Tandem DC generators produce conservative fields and the voltage can only be used once for acceleration. MULTI-TURN The Tandem van de Graaff is a clever to trick to use the voltage twice. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 13

14 A second history line Theory and proof-of-principle: 1924 Ising proposes time-varying fields across drift tubes. This is a true accelerator that can achieve energies above that given by the highest voltage in the system Wideröe demonstrates Ising s principle with a 1 MHz, 25 kv oscillator to make 50 kev potassium ions; the first linac. And on to a practical device: 1929 Lawrence, inspired by Wideröe and Ising, conceives the cyclotron; a coiled linac Livingston demonstrates the cyclotron by accelerating hydrogen ions to 80 kev Lawrence s cyclotron produces 1.25 MeV protons and he also splits the atom just a few weeks after Cockcroft & Walton. Lawrence received the Nobel Prize in CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 14

15 The players Gustaf Ising: Rolf Wideröe: Born 11/7/1902 in Oslo, Norway. Died Also contributed to the fields of power lines and cancer therapy. Ernest Orlando Lawrence: Born 8/8/1901 in South Dakota, USA. Third generation Norwegian. Died 27/8/1958. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 15

16 Livingston s demonstration cyclotron A glass envelope made from a flattened flask and silvered on the inside with a single Dee. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 16

17 Cyclotron Centripetal force F = evb = mv ρ 2 Constant revolution frequency f rev = v 2πρ = v eb 2π mv = eb 2πm Radius of gyration ρ = mv eb v +Ion F B ρ CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 17

18 One of Lawrence s cyclotrons Stanley Livingston and Ernest O. Lawrence (left to right) beside the 27 inch cyclotron at Berkeley circa The peculiar shape of the magnet s yoke arises from its conversion from a Poulson arc generator of RF current, formerly used in radio communication. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 18

19 The first true accelerator This principle is used in almost all of today s accelerators. The ions can reach energies above the highest voltage in the system. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 19

20 Leo Szilard was too late The first accelerator proposed by L. Szilard was a linac, appearing in a German patent application entitled "Acceleration of Corpuscles" and filed on 17 December The Figure shows the proposed layout. Though Szilard writes of "canal rays" in the patent application, he also refers to "corpuscles, e.g. ions or electrons." Considering the low-frequency RF sources available in those days, an apparatus of modest length would have worked only for rather heavy ions. Leo Szilard was a professional inventor. He dropped the above patent perhaps because of prior art by Ising. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 20

21 Wideröe s Linac Wideroe s first linac CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 21

22 Alvarez Linac Alvarez linac the first practical linac 32 MeV at Berkeley 1946: Particle gains energy at each gap. Drift tube lengths follow increasing velocity. The periodicity becomes regular as v c. His choice of 200 MHz became a de facto standard for many decades. Luis W. Alvarez was born in San Francisco, CA., on 13/6/1911. Died 1/9/1988. He received the Nobel physics prize in CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 22

23 And another history line, but fainter Also the birth of a true accelerator: 1923 Wideröe, a young Norwegian Ph.D. student draws in his laboratory notebook the design of the betatron with the well-known 2-to-1 rule. Two years later he adds the condition for radial stability, but does not publish in Aachen, Wideröe makes a model betatron, but it does not work. Discouraged he changes course and builds the world s first linac (see previous history line). All is quiet until Kerst re-invents the betatron and builds the first working machine for 2.2 MeV electrons (University of Illinois) Kerst also builds the world s largest betatron (300 MeV). CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 23

24 Wideröe s betatron Continuous acceleration betatron: Wideröe called this device a strahlung transformator because the beam effectively forms the secondary winding on a transformer. The above diagram is taken from his unpublished notebook (1923). This device is insensitive to relativistic effects and is therefore ideal for accelerating electrons. It is also robust and simple. The idea re-surfaced in 1940 with Kerst and Serber, who wrote a paper describing the beam oscillations. Subsequently the term betatron oscillation was adopted for these oscillations in all devices. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 24

25 Betatron CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 25

26 Classification by Maxwell Accelerators must use electric fields to transfer energy to/from an ion, because the force exerted by a magnetic field is always perpendicular to the motion. Mathematically speaking, the force exerted on an ion is: so that the rate at which work can be done on the ion is: but F ( v B) F = ee + e v = ee v + ( v B) v Each history line can be classified according to how the electric field is generated and used. e ( v B) v = 0. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 26

27 Use of the electric field E = - φ - A/ t Acceleration by DC voltages: Cockcroft & Walton rectifier generator Van de Graaff electrostatic generator Tandem electrostatic accelerator Acceleration by time-varying fields: E = - B/ t Betatron or unbunched acceleration B E Ion Resonant or bunched acceleration Linear accelerator (linac). Synchrotron. Cyclotron ( coiled linac). E B Ion CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 27

28 Status in 1940 Three acceleration methods had been exploited: DC voltage (e.g. Cockcroft and Walton), Resonant/bunched acceleration (e.g. cyclotron) Betatron/unbunched acceleration. Try to think of other possibilities for accelerating ions. * Progress now turns to applying these basic concepts more efficiently and to improving the technology. * This is an important question for the future. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 28

29 After 1940 in a nutshell 1943 Once again, Wideröe is a pioneer and patents colliding beams (pub. 1953) McMillan and Veksler independently propose synchronous acceleration with phase stability. They use an electron synchrotron, as example Goward and Barnes are first to make the synchrotron work in the UK Oliphant and Hyde start a 1 GeV machine in Birmingham, UK, but an American group overtakes them and is first with the 3 GeV Cosmotron at BNL Christofilos, and Courant, Livingston and Snyder independently invent strong focusing. CERN immediately drops its design for a weakfocusing, 10 GeV FFAG in favour of a strong-focusing, 28 GeV synchrotron MURA, US proposes particle stacking to increase beam intensity, opening the way for circular colliders. Trick Question: Why did McMillan receive the Nobel Prize and not Veksler? CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 29

30 Components of a synchrotron CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 30

31 More progress 1956 Tigner proposes linear colliders for high-energy electron machines AdA, an electron-positron storage ring is built in Frascati, Italy. This is the first single-ring, particle-antiparticle collider (first operated in Orsay, France) Budker and Skrinsky propose electron cooling Kapchinski & Teplyakov propose the RFQ (radiofrequency quadrupole) CERN operates the ISR protonproton collider. This is the first, particleparticle, intersecting-ring collider Blewett proposes the twin-bore superconducting magnet design. Now used in LHC van der Meer invents stochastic beam cooling opening the way for hadron, particle-antiparticle colliders The CERN ISR operates the first superconducting magnets (quads) to be used in a synchrotron ring. They are industrially built. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 31

32 And more 1982 CERN converts its SPS to a single-ring proton-antiproton collider C. Rubbia and S. van der Meer receive the Nobel physics prize for W & Z discoveries CERN starts LEP, the world s highest energy electron-positron collider HERA at DESY becomes the first major facility for colliding protons with electrons or positrons CERN runs superconducting rf cavities in LEP for physics RHIC at BNL becomes the world facility for colliding ions. 10 th September 2008 CERN starts the LHC, the world s highest energy protonproton collider (superconducting, twinbore dipoles). 20?? CERN has plans for a TeV linear collider, CLIC. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 32

33 Livingston chart 1 PeV FNAL 2x800 GeV Oct Beam energy or, for colliders, the equivalent beam energy on a fixed target 100 TeV 10 TeV 1 TeV 100 GeV 10 GeV P-p and p-pbar colliders ISR SPS AG proton synchrotrons Electron linacs Electron synchrotrons Synchrocyclotrons Proton linacs Sector-focused cyclotrons Tandems Year Bottom left corner, Milton Stanley Livingston s original chart from his book High energy accelerators CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 33

34 Classification of accelerators DC voltage generators: Cockcroft Walton generator. Van de Graaff. Tandem. Unbunched/continuous acceleration: Betatron. Betatron core. Bunched/resonant acceleration: RFQ. Linac. Cyclotron, synchrocyclotron. Microtron. FFAG (Fixed Field Alternating Gradient). Synchrotron. Colliders: Circular (single-ring, particle v anti-particle and intersecting-rings, particle v particle). Linear. Other classifications: Weak/strong focusing. Normal/superconducting magnets & cavities. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 34

35 In the beginning there was HEP, but more money now passes through non- HEP applications Accelerator applications: Synchrotron light sources. Spallation sources. Isotope production. Radiography. Cancer therapy. Ion implantation and surface metallurgy. Sterilisation. Proposed accelerator applications: Inertial fusion drivers. Nuclear incinerators. Rocket motors. Spin-offs from HEP and accelerators: PET scanners. NMR scanners. CAT scanners. Superconducting wires, cables and devices. Large-scale UHV systems. Large-scale cryogenic systems... CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 35

36 Where to next? Today s HEP accelerators are nearing practical limits. What can be done? 1982 ECFA held the first workshop of a series on advanced accelerating techniques Challenge of Ultra-high Energies New College, Oxford, UK. The goal was a new acceleration technique capable of reaching PeV energies and higher with equipment of a practical size. Four essential ingredients are: A new acceleration mechanism. Transverse stability. Longitudinal (phase) stability. Stability against collective effects. The candidates were: Plasma-beat-wave accelerator. Wake-field accelerator. Lasers with gratings. Lasers on dense bunches. But the search is still on for a new HEP accelerator. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 36

37 How far is beyond? The CERN LHC will operate 2 7 TeV (1 TeV ev) beams in head-on collision. Only cosmic rays provide a glimpse of what lies beyond. The cosmic ray spectrum is expected to extend up to the Planck energy ( ev about times higher than the LHC), above which the universe is thought to be opaque. The Planck energy is the order of magnitude expected for the energy of a vibrating string in string theory. The Planck energy is roughly 2 billion joules, the energy supplied when tanking up a car. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 37

38 A closer look at cyclotrons Cyclotrons started in HEP, but today they are important for their industrial and medical applications. The cyclotron s success is due to its robust and compact design with adequate intensity and quasi-continuous beam. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 38

39 Cyclotron road map 1932 Lawrence s first cyclotron works 1930s Studies on neutron therapy using a cyclotron Early 1990s Superconducting cyclotron mounted directly on gantry for neutron therapy 2000s IBA propos a superconducting machine for 400 MeV/u carbon ions. This machine could displace synchrotrons and take the world market. 1980s IBA s Cyclone 30 becomes the de facto standard for isotope production Late 1990s Cyclotron establishes itself as the protontherapy standard with passive spreading CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 39

40 Evolution of cyclotrons Beam power [kw] s -Cyclone 30 upgrade 1-2 ma -Superconducting cyclotrons Current [µa] External, multi-cusp, H beam source. With axial injection and deep valley magnet design 500 µa at 30 MeV (IBA Cyclone 30) s H beam from internal PIG source gives variable energy and multi-porting, but poor vacuum. ~200 µa at 42 MeV (TCC Berkeley) s to 1970s Classical isochronous cyclotrons using extraction by electrostatic deflection are limited by heating. 200 µa at 10 MeV CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 40

41 Cyclones 30 and 235 (courtesy IBA) CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 41

42 Gantry-mounted sc cyclotron Gantry-mounted, superconducting, deuteron cyclotron for neutron therapy Harper Hospital, Detroit H. Blosser et al, Hadrontherapy in Onc, 1994 CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 42

43 Recognising synchrotron lattices Can we recognise the types of lattice and guess the application of a synchrotron from its lattice design? Lets look at typical examples: Early accelerators for physics More recent accelerators for physics Spallation sources Synchrotron light sources Cancer therapy machines CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 43

44 An early synchrotron 3 GeV proton synchrotron Saturne at Saclay. A Van de Graaff injector lies out of view front-right. The magnet structure is quasi-continuous because the designers were not skilled in the design of long drift spaces. These machines are invariably plain accelerators for physics research with the experiments external to the machine. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 44

45 A more advanced AG accelerator Uses a basic FODO cell with the F and D quads split into 2 units. Between the split quadrupoles, the betatron amplitude functions are quasi constant. The dipoles are placed between the F quads to have minimum vertical beam size (i.e. min. cost). However, the drift spaces are still short. ADONE CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 45

46 Controlling dispersion The rings shown so far simply repeat a standard cell n times to reach 2π of bending. This works for plain accelerators and often leads to an economical solution in which all quadrupoles for example are powered by a single power converter. In more advanced lattices, we would like to have regions with zero dispersion e.g. in RF cavities. This is done in small rings by closing the dispersion in bumps. To close a dispersion bump one needs 180 to 360 of phase advance in the plane of bending. This leads to solutions for rings with two, or three or four or more closed dispersion bumps separated by dispersion-free sections. Each closed bump forms a corner and the ring looks triangular or square or pentagonal and so on. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 46

47 A triangular ring using a triplet A triplet is another possible cell for a ring. In this case, the large horizontal phase advance at the centre of the triplet is used to make 3 closed dispersion bumps. The waist in the vertical betatron amplitude in long straight sections is used for the dipoles. This keeps the aperture requirements and cost down. AUSTRON Spallation source CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 47

48 H minus stripping - a special feature of spallation sources Inject H minus Unstripped H minus Partially stripped H 0 Main dipoles Weak dipoles Majority of beam continues on central orbit AUSTRON Spallation source CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 48

49 Light source lattice Chasman-Greene, double-bend achromat, highbrightness lattice. The aim is to minimise D x (s) and β x (s) in the dipoles. Each cell supports a closed dispersion bump. There are 4 bumps making a square ring. NSLS Light source CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 49

50 A medical synchrotron The PIMMS medical synchrotron is an example of a lattice customised for a particular use. Injection and extraction use electrostatic septa for quasi continuous operation. The long straight sections have zero dispersion for rf cavities and minimum beam size at injection/extraction. Phase advances are designed for the slow extraction. PIMMS medical ring CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 50

51 Large rings Large rings, such as the LHC, often have a basic FODO cell in the arcs. The overall ring has an n-fold symmetry containing the n- arcs and n straight regions in which the physics experiments are mounted. Between the arcs and the straight regions there is the socalled dispersion suppressor that brings the dispersion function to zero in the straight region in a controlled way. There are several schemes for dispersion suppressors (see one example on next slide). The straight regions contain the injection and extraction and the RF cavities, which, in an electron machine like LEP, can occupy hundreds of metres. A dispersion-free straight region may also contains a lowβ insertion for physics. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 51

52 Missing-magnet suppressor Lattice functions of missing-magnet suppressor for a 60 FODO cell. Note how β x and β z hardly notice the suppression of D x. Arc µ x =60 2 missing dipoles Zero dispersion straight section CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 52

53 The FFAG FFAG = Fixed Field Alternating Gradient FFAGs are experiencing a rebirth, since their conception in the 1950s. One could describe them as a solution looking for an application. FFAGs have a large momentum acceptance when considered as a single aperture fixed-field magnet, but a small momentum range when considered as an accelerator. Fast cycling synchrotrons can approach a range of 1:10, while slow ramping synchrotrons can approach a ratio of 1:20, whereas FFAGs may approach a range of 1:5. CAS_12- P.J. Bryant - History and Applications of Accelerators - 2 lectures - Slide 53